Skip to content

Observability

  • webhook.handler.failed
  • conversation.work.failed
  • conversation.work.recovery.failed
  • agent.continue.schedule.failed
  • agent.turn.failed
  • agent.turn.timed_out
  • agent.tool_call.failed
  • http.server.request
  • queue.enqueue_message
  • queue.process_message
  • chat.turn
  • gen_ai.invoke_agent
  • gen_ai.execute_tool
  • http.response.status_code
  • url.path
  • app.queue.delivery_count
  • messaging.message.conversation_id
  • gen_ai.request.model
  • error.type
event.name:webhook.handler.failed
event.name:conversation.work.failed OR event.name:conversation.work.recovery.failed
event.name:conversation.work.pending.requeued OR event.name:conversation.work.lease_expired.requeued
event.name:agent.continue.schedule.failed OR event.name:agent.continue.lock.busy
event.name:agent.turn.timed_out OR event.name:agent.turn.failed
event.name:agent.tool_call.failed

Use symptom-driven playbooks in Reliability Runbooks.